Learning Device for FMEA Sheet Creation Using Past Case Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional FMEA sheet creation support systems fail to effectively utilize the structural features of FMEA sheets and past FMEA sheets, leading to incomplete information provision to users.
Innovation Solution
A learning device that generates correspondence relationship training data and models by integrating operation process information and risk sentences from past FMEA sheets, enabling the system to learn and provide more comprehensive information during FMEA sheet creation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If conventional support systems create feature quantities solely from text information on FMEA sheets, then the system can provide basic text search functionality, but the system fails to provide useful information by not considering structural features and past FMEA sheets
Solution Approach 1:
The patent merges multiple information sources (text information, structural features of FMEA sheets, and past FMEA sheets) to create comprehensive feature quantities. This combining approach ensures that the system utilizes all available information including the sheet structure with rows and columns, operation process information, risk sentences, and historical data, thereby preventing loss of useful information while maintaining a unified analysis framework.
Solution Approach 2:
The patent performs preliminary extraction and organization of structural features and correspondence relationships from past FMEA sheets before conducting new analyses. By pre-processing and storing the structural characteristics (rows, columns, operation processes, risk sentences) and their correspondence relationships in advance, the system prepares comprehensive feature quantities that can be efficiently utilized during actual FMEA sheet creation without losing important information.
2Reliability
If the system considers structural features and past FMEA sheets, then the system can provide more useful information, but the processing complexity and data requirements increase
Solution Approach 1:
The patent segments the FMEA sheet structure into distinct components (rows representing operation processes, columns representing risk information, and specific fields like operation process information and risk sentences). This segmentation allows the system to process and analyze each structural element separately, extracting feature quantities from specific segments while maintaining the overall structural context, thereby improving reliability without overwhelming system complexity.
Solution Approach 2:
The patent introduces correspondence relationship models as intermediaries that connect operation process information with risk sentences based on learned relationships from past FMEA sheets. These models act as mediators that translate structural features and historical data into meaningful associations, enhancing the accuracy and reliability of FMEA sheet creation by providing structured connections between different elements rather than processing raw data directly.
3Adaptability or versatility
If conventional systems rely on operator knowledge and experience, then the creation process can be simple, but the FMEA sheet may miss failures that the operator has not experienced
Solution Approach 1:
The patent implements feedback mechanisms by analyzing correspondence relationships between operation processes and risk sentences from past FMEA sheets. The system learns from historical data and provides feedback in the form of suggested risk sentences and failure modes during new FMEA sheet creation. This feedback loop enables the system to suggest failures that operators might have missed based on their limited experience, thereby improving completeness of failure identification.
Solution Approach 2:
The patent utilizes copying by extracting and reusing correspondence relationships and risk sentence patterns from past FMEA sheets. Instead of relying solely on operator knowledge, the system copies proven failure modes and risk associations from historical documents and applies them to new FMEA analyses. This approach expands the operator's effective experience base and improves the completeness of failure identification without requiring additional time for manual research.
Data Source
AI summary
A device includes: a storage unit that stores a past case sheet created in the past as a management sheet that includes rows, each of which includes at least operation process information indicating one operation process and a risk sentence indicating information about a risk in the one operation process; a training data generating unit that generates correspondence relationship training data, which includes a positive example and a negative example, the positive example being a combination of the operation process information included in one of the rows in the past case sheet and the risk sentence included in the one row, the negative example being a combination of the operation process information included in the one row and the risk sentence included in a row different from the one row; and a correspondence relationship learning unit that generates a correspondence relationship model by using the correspondence relationship training data.